Prevalence and Clinical Outcomes of Respiratory Syncytial Virus vs Influenza in Adults Hospitalized With Acute Respiratory Illness From a Prospective Multicenter Study

BACKGROUND

Current understanding of severe respiratory syncytial virus (RSV) infections in adults is limited by clinical underrecognition. We compared the prevalence, clinical characteristics, and outcomes of RSV infections vs influenza in adults hospitalized with acute respiratory illnesses (ARIs) in a prospective national surveillance network.

METHODS

Hospitalized adults who met a standardized ARI case definition were prospectively enrolled across 3 respiratory seasons from hospitals participating across all sites of the US Hospitalized Adult Influenza Vaccine Effectiveness Network (2016-2019). All participants were tested for RSV and influenza using real-time reverse-transcription polymerase chain reaction assay. Multivariable logistic regression was used to test associations between laboratory-confirmed infection and characteristics and clinical outcomes.

RESULTS

Among 10 311 hospitalized adults, 6% tested positive for RSV (n = 622), 18.8% for influenza (n = 1940), and 75.1% negative for RSV and influenza (n = 7749). Congestive heart failure (CHF) or chronic obstructive pulmonary disease (COPD) was more frequent with RSV than influenza (CHF: 37.3% vs 28.8%, P < .0001; COPD: 47.6% vs 35.8%, P < .0001). Patients with RSV more frequently had longer admissions (odds ratio [OR], 1.38; 95% confidence interval [CI], 1.06-1.80) for stays >1 week) and mechanical ventilation (OR, 1.45; 95% CI, 1.09-1.93) compared with influenza but not compared with the influenza-negative group (OR, 1.03; 95% CI, .82-1.28 and OR, 1.17; 95% CI, .91-1.49, respectively).

CONCLUSIONS

The prevalence of RSV across 3 seasons was considerable. Our findings suggest that those with RSV have worse outcomes compared with influenza and frequently have cardiopulmonary conditions. This study informs future vaccination strategies and underscores a need for RSV surveillance among adults with severe ARI.

Effectiveness of Monovalent mRNA COVID-19 Vaccination in Preventing COVID-19-Associated Invasive Mechanical Ventilation and Death Among Immunocompetent Adults During the Omicron Variant Period - IVY Network, 19 U.S. States, February 1, 2022-January 31, 2023

As of April 2023, the COVID-19 pandemic has resulted in 1.1 million deaths in the United States, with approximately 75% of deaths occurring among adults aged ≥65 years (1). Data on the durability of protection provided by monovalent mRNA COVID-19 vaccination against critical outcomes of COVID-19 are limited beyond the Omicron BA.1 lineage period (December 26, 2021-March 26, 2022). In this case-control analysis, the effectiveness of 2-4 monovalent mRNA COVID-19 vaccine doses was evaluated against COVID-19-associated invasive mechanical ventilation (IMV) and in-hospital death among immunocompetent adults aged ≥18 years during February 1, 2022-January 31, 2023. Vaccine effectiveness (VE) against IMV and in-hospital death was 62% among adults aged ≥18 years and 69% among those aged ≥65 years. When stratified by time since last dose, VE was 76% at 7-179 days, 54% at 180-364 days, and 56% at ≥365 days. Monovalent mRNA COVID-19 vaccination provided substantial, durable protection against IMV and in-hospital death among adults during the Omicron variant period. All adults should remain up to date with recommended COVID-19 vaccination to prevent critical COVID-19-associated outcomes.

K-medoids clustering of hospital admission characteristics to classify severity of influenza virus infection

Background

Patients are admitted to the hospital for respiratory illness at different stages of their disease course. It is important to appropriately analyse this heterogeneity in surveillance data to accurately measure disease severity among those hospitalized. The purpose of this study was to determine if unique baseline clusters of influenza patients exist and to examine the association between cluster membership and in-hospital outcomes.

Methods

Patients hospitalized with influenza at two hospitals in Southeast Michigan during the 2017/2018 (n = 242) and 2018/2019 (n = 115) influenza seasons were included. Physiologic and laboratory variables were collected for the first 24 h of the hospital stay. K-medoids clustering was used to determine groups of individuals based on these values. Multivariable linear regression or Firth's logistic regression were used to examine the association between cluster membership and clinical outcomes.

Results

Three clusters were selected for 2017/2018, mainly differentiated by blood glucose level. After adjustment, those in C₁₇1 had 5.6 times the odds of mechanical ventilator use than those in C₁₇2 (95% CI: 1.49, 21.1) and a significantly longer mean hospital length of stay than those in both C₁₇2 (mean 1.5 days longer, 95% CI: 0.2, 2.7) and C₁₇3 (mean 1.4 days longer, 95% CI: 0.3, 2.5). Similar results were seen between the two clusters selected for 2018/2019.

Conclusion

In this study of hospitalized influenza patients, we show that distinct clusters with higher disease acuity can be identified and could be targeted for evaluations of vaccine and influenza antiviral effectiveness against disease attenuation. The association of higher disease acuity with glucose level merits evaluation.

Management of hepatocellular carcinoma from diagnosis in routine clinical practice

Aim: To assess real-world management of patients diagnosed with hepatocellular carcinoma (HCC) within an integrated delivery network. Materials & methods: A retrospective cohort analysis of adults newly diagnosed with HCC from January 2014 to March 2019. Overall survival and treatment journey were assessed over the entire available follow-up period per patient. Results: Of the 462 patients, 85% had ≥1 treatment. The 24-month overall survival rate (95% CI) from first treatment was 77% (72–82%). Majority of Child-Pugh class A (71%) and B (60%) patients received locoregional therapy first. Half (53.6%) of the patients with liver transplantation first were Child-Pugh class C patients. Sorafenib was the predominant systemic therapy. Conclusion: This integrated delivery network data analysis offers a comprehensive insight into the real-world management of HCC.

Vaccine effectiveness against COVID-19 among symptomatic persons aged ≥12 years with reported contact with COVID-19 cases, February-September 2021

BACKGROUND

Individuals in contact with persons with COVID-19 are at high risk of developing COVID-19; protection offered by COVID-19 vaccines in the context of known exposure is poorly understood.

METHODS

Symptomatic outpatients aged ≥12 years reporting acute onset of COVID-19-like illness and tested for SARS-CoV-2 between February 1 and September 30, 2021 were enrolled. Participants were stratified by self-report of having known contact with a COVID-19 case in the 14 days prior to illness onset. Vaccine effectiveness was evaluated using the test-negative study design and multivariable logistic regression.

RESULTS

Among 2229 participants, 283/451 (63%) of those reporting contact and 331/1778 (19%) without known contact tested SARS-CoV-2-positive. Adjusted vaccine effectiveness was 71% (95% confidence interval [CI], 49%-83%) among fully vaccinated participants reporting a known contact versus 80% (95% CI, 72%-86%) among those with no known contact (p-value for interaction = 0.2).

CONCLUSIONS

This study contributes to growing evidence of the benefits of vaccinations in preventing COVID-19 and support vaccination recommendations and the importance of efforts to increase vaccination coverage.

Incidence and economic burden of respiratory syncytial virus among adults in the United States: A retrospective analysis using 2 insurance claims databases

BACKGROUND: Respiratory syncytial virus (RSV) is a common, contagious, and seasonal pathogen causing 64 million acute respiratory infections annually in adults and children worldwide. High-risk adults, including older adults and those with cardiopulmonary conditions or weakened immune systems, are more likely to be infected. However, limited information exists on RSV incidence and associated costs among adults, including high-risk patients. OBJECTIVE: To evaluate the annual incidence of medically attended, International Classification of Diseases (ICD)-coded RSV among commercially insured adults and assess health care costs among adults with ICD-coded RSV in the United States. METHODS: Optum's deidentified Clinformatics Data Mart Database (January 01, 2007, to June 30, 2020) and IBM's MarketScan Databases (January 01, 2000, to July 31, 2020) were used. Medically attended, ICD-coded RSV incidence among adults was assessed from July 1 of a given year to June 30 of the next year and reported per 100,000 population. Trends in all-cause mean weekly costs pre-RSV and post-RSV diagnosis were reported. Results were reported overall and among patients aged 60-64 years, 65 years or older, 85 years or older, and 18-59 years at high risk of severe RSV (defined as having cardiopulmonary conditions or a weakened immune system). RESULTS: Annual incidence of medically attended, ICD-coded RSV in adults overall was 22.0-52.9 in Optum and 23.4-63.6 in MarketScan. Incidence rates were higher among patients aged 60-64 years (Optum: 25.2-66.1; MarketScan: 31.9-82.1), 65 years or older (Optum: 37.3-75.5; MarketScan: 54.1-97.3), 85 years or older (Optum: 92.4-140.6; MarketScan: 79.4-234.7), and 18-59 years at high risk of severe RSV (Optum: 41.3-135.9; MarketScan: 46.3-112.4). Mean weekly costs increased during the week before (Optum: $2,325; MarketScan: $2,080) and post-RSV diagnosis (Optum: $9,523; MarketScan: $3,551), compared with those in weeks 2-8 pre-RSV diagnosis (Optum: $1,350; MarketScan: $872). The increases in mean weekly costs during the week before and the week following RSV diagnosis were higher among patients aged 60-64 years (mean weekly costs in weeks 2-8 pre-RSV, week 1 pre-RSV, week 1 post-RSV; Optum: $1,623, $2,690, $10,823; MarketScan: $1,259, $2,992, $5,069), 65 years or older (Optum: $1,731, $3,067, $12,866; MarketScan: $1,517, $3,571, $5,268), 85 years or older (Optum: $1,563, $2,430, $18,134; MarketScan: $1,613, $4,113, $6,231), and 18-59 years at high risk of severe RSV (only for MarketScan: $1,237, $3,294, $5,531; costs were similar for Optum). CONCLUSIONS: Incidence of medically attended, ICD-coded RSV in adults was 22.0-63.6 per 100,000 population, a likely underestimation since RSV was not systematically tested and only RSV-coded cases were observed. Incremental costs associated with RSV were substantial. Incidence rates and costs were higher among patients aged 60 years or older and patients at high risk of severe RSV. DISCLOSURES: This study was sponsored by Janssen Scientific Affairs, LLC. The sponsor was involved in the study design, interpretation of results, manuscript preparation, and publication decisions. B. Brookhart and D. Anderson are employees of Janssen Scientific Affairs, LLC, and are stockholders of Johnson & Johnson. C. Rossi, B. Emond, J. Wang, P. Lefebvre, and M.-H. Lafeuille are employees of Analysis Group, Inc., a consulting company that has provided paid consulting services to Janssen Scientific Affairs, LLC, which funded the development and conduct of this study and manuscript. M. Mesa-Frias. and S. Drummond are former employees of Janssen Scientific Affairs, LLC. L. Lamerato is an employee of Henry Ford Health System and received research funding from Janssen Scientific Affairs, LLC.

Messenger RNA Vaccine Effectiveness Against Coronavirus Disease 2019 Among Symptomatic Outpatients Aged ≥16 Years in the United States, February-May 2021

Evaluations of vaccine effectiveness (VE) are important to monitor as coronavirus disease 2019 (COVID-19) vaccines are introduced in the general population. Research staff enrolled symptomatic participants seeking outpatient medical care for COVID-19-like illness or severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) testing from a multisite network. VE was evaluated using the test-negative design. Among 236 SARS-CoV-2 nucleic acid amplification test-positive and 576 test-negative participants aged ≥16 years, the VE of messenger RNA vaccines against COVID-19 was 91% (95% confidence interval, 83%-95%) for full vaccination and 75% (55%-87%) for partial vaccination. Vaccination was associated with prevention of most COVID-19 cases among people seeking outpatient medical care.

Patient Experience of Interpersonal Processes of Care and Subsequent Utilization of Hormone Therapy for Non-Metastatic Breast Cancer

Purpose: We examined the association between patient experience of care and utilization of hormone therapy (HT) in the treatment of non-metastatic hormone-receptor positive (HR+) breast cancer. Methods: Patients newly diagnosed with non-metastatic breast cancer were recruited from 2006 to 2010 for a longitudinal multisite cohort study in New York, NY, Detroit, MI, and Northern California. Of 1,145 patients surveyed, 797 had HR+ tumors eligible for HT and all necessary data. We assessed patient experience 4 to 8 weeks after recruitment using 6 subscales of the Interpersonal Processes of Care (IPC) survey: compassion, discrimination, and hurriedness in communication, as well as concern elicitation, result explanation, and patient-centered decision-making. Subscales ranged from 1 to 5 where higher values indicated better experiences. HT for 5 years is standard care for HR+ breast cancer so utilization was defined as time from diagnosis to HT initiation and time from HT initiation to early discontinuation before 5 years as calculated from follow-up survey responses. We evaluated the relationship between patient experience and utilization using Cox proportional hazard models, controlling for education, income, insurance, marital status, social support, site of care, age at diagnosis, stage, grade, tumor size, Charlson comorbidity index, and chemotherapy. Results: Median age at diagnosis was 59 years (interquartile range 51–66) with the majority diagnosed at clinical stage 1 (54%) and with low or moderate grade disease (78%). Less hurried communication was associated with increased probability of HT initiation (Hazards Ratio (HR) 1.15; 95% Confidence Interval (CI) 1.03, 1.30; p = 0.018). Conversely, more patient-centered decision-making was associated with increased probability of early discontinuation (HR 1.29; CI, 1.03, 1.63; p = 0.028). All other associations were null. Conclusion: While unhurried communication was associated with initiation of hormone therapy, patient-centered decision-making was associated with early discontinuation. Different aspects of patient experience may have vastly different relationships with patient utilization of health services. Actionable assessments of patient experience may require measurement along multiple dimensions.

Fitness and prostate cancer screening, incidence, and mortality: Results from the Henry Ford Exercise Testing (FIT) Project

BACKGROUND

The relation between cardiorespiratory fitness (CRF) and prostate cancer is not well established. The objective of this study was to determine whether CRF is associated with prostate cancer screening, incidence, or mortality.

METHODS

The Henry Ford Exercise Testing Project is a retrospective cohort study of men aged 40 to 70 years without cancer who underwent physician-referred exercise stress testing from 1995 to 2009. CRF was quantified in metabolic equivalents of task (METs) (<6 [reference], 6-9, 10-11, and ≥12 METs), estimated from the peak workload achieved during a symptom-limited, maximal exercise stress test. Prostate-specific antigen (PSA) testing, incident prostate cancer, and all-cause mortality were analyzed with multivariable adjusted Poisson regression and Cox proportional hazard models.

RESULTS

In total, 22,827 men were included, of whom 739 developed prostate cancer, with a median follow-up of 7.5 years. Men who had high fitness (≥12 METs) had an 28% higher risk of PSA screening (95% CI, 1.2-1.3) compared with those who had low fitness (<6 METs. After adjusting for PSA screening, fitness was associated with higher prostate cancer incidence (men aged <55 years, P = .02; men aged >55 years, P ≤ .01), but not with advanced prostate cancer. Among the men who were diagnosed with prostate cancer, high fitness was associated with a 60% lower risk of all-cause mortality (95% CI, 0.2-0.9).

CONCLUSIONS

Although men with high fitness are more likely to undergo PSA screening, this does not fully account for the increased incidence of prostate cancer seen among these individuals. However, men with high fitness have a lower risk of death after a prostate cancer diagnosis, suggesting that the cancers identified may be low-risk with little impact on long-term outcomes.

Abstract 4348: Cardiorespiratory fitness levels in men who develop prostate cancer and its association with all-cause mortality

Background: Cardiorespiratory fitness (CRF) is associated with a reduction in both cancer- and cardiovascular-specific mortality. In men however, little is known about the impact of pre-diagnostic CRF on mortality after prostate cancer. We hypothesized that among men with prostate cancer, low levels of CRF, measured prior to diagnosis, are associated with a higher risk of all-cause mortality.

Methods: From the Henry Ford FIT Project, a cohort of 37,730 men who underwent clinically-indicated exercise stress test from 1991 through 2009, we identified 1,440 men who developed prostate cancer after stress test. CRF was measured in peak metabolic equivalents of task (METs) and categorized as less than 6, 6 to 9, 10 to 11, and at least 12 (reference). Multivariable Cox proportional hazard models were developed to evaluate the association between CRF, measured prior to prostate cancer diagnosis, and all-cause mortality. Models were adjusted for age at prostate cancer diagnosis (baseline), race, BMI, current aspirin and statin use, smoking history, hypertension, diabetes, prior cardiovascular disease (myocardial infarction and heart failure), time from stress test to prostate cancer diagnosis, and cancer stage (local, regional, distant) at diagnosis.

Results: Mean age at time of stress test was 61 +/- 9 yr (57% white, 39% black), with 7 (+/-4) yrs of follow up after prostate cancer diagnosis. 81% of men were diagnosed with localized prostate cancer; 15% with regional disease, and 2% with distant metastatic disease. Mean time from stress test to diagnosis was 6 +/- 5 yrs. Mean METs achieved was 9 (+/- 3). Patients with low fitness (METs less than 6) before diagnosis had 2.6 times the risk of all-cause mortality (95% CI 1.6-4.2) compared to those with high fitness (METS at least 12). Those with CRF of 6-9 METs had 1.9 times the risk (95% CI 1.2-3.0). The risk of mortality was not different between the two highest fitness groups (P= 0.71; P for trend across all fitness groups less than 0.01). Results were similar when stratified by race.

Conclusion: Low CRF prior to the diagnosis of prostate cancer in men is associated with a higher risk of all-cause mortality. This study highlights the prognostic importance of fitness even before a cancer diagnosis.

Citation Format: Cara Reiter-Brennan, Omar Dzaye, Mouaz H. Al-Mallah, Zeina Dardari, Clinton A. Brawner, Lois E. Lamerato, Steven J. Keteyian, Jonathan K. Ehrman, Kala Visvanathan, Michael J. Blaha, Catherine Handy Marshall. Cardiorespiratory fitness levels in men who develop prostate cancer and its association with all-cause mortality [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 4348.

Abstract 5773: Cardiorespiratory fitness and PSA screening patterns in the Henry Ford FIT Project

Background: High levels of cardiorespiratory fitness (CRF) have been associated with a decreased risk of many cancers, but the association with prostate cancer (PCa) is less clear. A possible explanation for the lack of consistency is that prostate specific antigen (PSA) screening and in turn prostate cancer detection is more frequent in men with higher fitness although the relationship between fitness and PSA screening is not established.

Methods: The Henry Ford (HF) FIT Project is a retrospective cohort study of 69,894 consecutive patients who underwent physician-referred exercise stress testing from 1991 through 2009. Cancer diagnosis was identified through linkage to the HF tumor registry. We included men aged 40-70 yr who had CRF testing beginning in 1995 (when PSA screening became widespread), without prevalent cancer, followed at least 3 yrs and up to 10 years, who never develop prostate cancer (n=12,442). CRF was measured in metabolic equivalents of task (METs) and categorized as less than 6 (ref), 6-9, 10-11, and at least 12. PSA values were abstracted from the HF electronic medical record between 1995 and 2010. Individuals with at least 3 screening PSA tests were categorized as high screeners. Logistic regression was used to evaluate the relationship between CRF and high PSA screeners, adjusted for age at stress test and race.

Results: Participants had a mean age of 54 ± 8 yrs (67% white; 25% black) and mean follow-up was 7±2 yrs. Compared to individuals with low fitness (less than 6 METs, reference), those with higher fitness (6-9, 10-11, at least 12 METs) had higher odds of being a high PSA screener (Table). When stratified by race, the OR for black patients were not statistically different than white patients (Table). Results were similar among those with comorbidities (myocardial infarction, heart failure, or diabetes) at baseline.

Odds ratio of being a high screener (at least 3 PSA tests) in individuals followed 3-10 yearsMETs categoryAll (n=12,442Whites (n=8,356Blacks (n=3,049)Patients w/comorbidities (n=3,924)&lt;6RefRefRefRef6-91.6(1.3-1.8)1.6(1.3-2.0)1.4(1.1-1.9)1.6(1.3-2.0)10-111.9(1.6-2.2)1.9(1.6-2.4)1.7(1.3-2.2)2.0(1.7-2.4)&gt;=122.4 (2.1-2.9)2.5(2.0-3.1)2.1(1.6-2.8)3.2(2.5-4.0)

Conclusion: High CRF is associated with more frequent PSA screening. This points towards a healthy screening bias and should be accounted for in studies looking at fitness and incident prostate cancer.

Citation Format: Cara Reiter-Brennan, Omar Dzaye, Mouaz H. Al-Mallah, Zeina Dardari, Clinton A. Brawner, Lois E. Lamerato, Steven J. Keteyian, Jonathan K. Ehrman, Kala Visvanathan, Michael J. Blaha, Catherine Handy Marshall. Cardiorespiratory fitness and PSA screening patterns in the Henry Ford FIT Project [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 5773.

The association of fitness and body mass index (BMI) on all-cause mortality in cancer survivors: The Henry Ford Exercise Testing Project (The FIT Project)

7060

Background: The obesity paradox –i.e. inverse associations between body mass index (BMI) and mortality – has been reported in patients with cancer, heart failure, and diabetes. However, the influence of cardiorespiratory fitness (CRF) on this relationship is not well established. This study assesses the association of BMI and CRF with all-cause mortality among cancer patients. Methods: The Henry Ford (HF) FIT Project is a retrospective cohort study of 69,885 consecutive patients who underwent physician-referred exercise stress testing from 1991 through 2009. Cancer diagnosis was identified through linkage to the HF tumor registry. We included patients 40-70 years old, with BMI recorded, at time of exercise test, with a history of cancer > 6 months prior. BMI was categorized as normal (18.5-24.9kg/m2), overweight (25-29.9kg/m2), or obese ( > = 30kg/m2). All-cause mortality was obtained from the National Death Index. Because of a significant interaction between BMI and cancer type, patients with breast or prostate cancer were excluded. Multivariable adjusted Cox proportional hazard models were used to evaluate the association of CRF andBMI with all-cause mortality; adjusted for age at exercise test, sex, diabetes, smoking, cancer stage, and time from cancer diagnosis to exercise test. Results: Included were 676 patients with a mean age of 58 years (SD 7.5), 51% female, 70% White, 25% Black, with a median of 4.8 years from diagnosis to exercise test and median follow up time of 10.3 years. Among patients achieving < 10 METs, those who are overweight and obese had a lower risk of mortality HR 0.47 (95% CI 0.25,0.86) and HR 0.44 (95% CI 0.26, 0.74, respectively), compared to those with normal BMI. Among patients with METs > = 10, those who were overweight had the lowest risk of all-cause mortality (HR 0.23, 95% CI 0.09-0.62) compared to normal weight, while no statistically significant different risk of mortality was observed when comparing those who are obese to normal weight (HR 0.37, 95% CI 0.13-1.06). In an analysis combining BMI and fitness groups (four categories), those with BMI > = 25 and METs > = 10 had the lowest risk of all-cause mortality (Table). Conclusions: In non-breast/non-prostate cancer patients, increased BMI is associated with improved overall survival in those with METs < 10, while a U-shaped relationship between BMI and all-cause mortality exists among those with METs > = 10. [Table: see text]

Influenza Vaccine Effectiveness in the Inpatient Setting: Evaluation of Potential Bias in the Test-Negative Design by Use of Alternate Control Groups

The test-negative design is validated in outpatient, but not inpatient, studies of influenza vaccine effectiveness. The prevalence of chronic pulmonary disease among inpatients can lead to nonrepresentative controls. Test-negative design estimates are biased if vaccine administration is associated with incidence of noninfluenza viruses. We evaluated whether control group selection and effects of vaccination on noninfluenza viruses biased vaccine effectiveness in our study. Subjects were enrolled at the University of Michigan and Henry Ford hospitals during the 2014-2015 and 2015-2016 influenza seasons. Patients presenting with acute respiratory infection were enrolled and tested for respiratory viruses. Vaccine effectiveness was estimated using 3 control groups: negative for influenza, positive for other respiratory virus, and pan-negative individuals; it was also estimated for other common respiratory viruses. In 2014-2015, vaccine effectiveness was 41.1% (95% CI: 1.7, 64.7) using influenza-negative controls, 24.5% (95% CI: -42.6, 60.1) using controls positive for other virus, and 45.8% (95% CI: 5.7, 68.9) using pan-negative controls. In 2015-2016, vaccine effectiveness was 68.7% (95% CI: 44.6, 82.5) using influenza-negative controls, 63.1% (95% CI: 25.0, 82.2) using controls positive for other virus, and 71.1% (95% CI: 46.2, 84.8) using pan-negative controls. Vaccination did not alter odds of other respiratory viruses. Results support use of the test-negative design among inpatients.

Cardiorespiratory fitness and incident lung and colorectal cancer in men and women: Results from the Henry Ford Exercise Testing (FIT) cohort

BACKGROUND

To the authors' knowledge, the relationship between cardiorespiratory fitness (CRF) and lung and colorectal cancer outcomes is not well established.

METHODS

A retrospective cohort study was performed of 49,143 consecutive patients who underwent clinician-referred exercise stress testing from 1991 through 2009. The patients ranged in age from 40 to 70 years, were without cancer, and were treated within the Henry Ford Health System in Detroit, Michigan. CRF, measured in metabolic equivalents of task (METs), was categorized as <6 (reference), 6 to 9, 10 to 11, and ≥12. Incident cancer was obtained through linkage to the cancer registry and all-cause mortality from the National Death Index.

RESULTS

Participants had a mean age of 54 ± 8 years. Approximately 46% were female, 64% were white, 29% were black, and 1% were Hispanic. The median follow-up was 7.7 years. Cox proportional hazard models, adjusted for age, race, sex, body mass index, smoking history, and diabetes, found that those in the highest fitness category (METs ≥12) had a 77% decreased risk of lung cancer (hazard ratio [HR], 0.23; 95% CI, 0.14-0.36) and a 61% decreased risk of incident colorectal cancer (HR, 0.39; 95% CI, 0.23-0.66; with additional adjustment for aspirin and statin use). Among those diagnosed with lung and colorectal cancer, those with high fitness had a decreased risk of subsequent death of 44% and 89%, respectively (HR, 0.56 [95% CI, 0.32-1.00] and HR, 0.11 [95% CI, 0.03-0.37], respectively).

CONCLUSIONS

In what to the authors' knowledge is the largest study performed to date, higher CRF was associated with a lower risk of incident lung and colorectal cancer in men and women and a lower risk of all-cause mortality among those diagnosed with lung or colorectal cancer.

Evaluation of correlates of protection against influenza A(H3N2) and A(H1N1)pdm09 infection: Applications to the hospitalized patient population

BACKGROUND

Influenza vaccines are important for prevention of influenza-associated hospitalization. However, the effectiveness of influenza vaccines can vary by year and influenza type and subtype and mechanisms underlying this variation are incompletely understood. Assessments of serologic correlates of protection can support interpretation of influenza vaccine effectiveness in hospitalized populations.

METHODS

We enrolled adults hospitalized for treatment of acute respiratory illnesses during the 2014-2015 and 2015-2016 influenza seasons whose symptoms began <10 days prior to enrollment. Influenza infection status was determined by RT-PCR. Influenza vaccination status was defined by self-report and medical record/registry documentation. Serum specimens collected at hospital admission were tested in hemagglutination-inhibition (HAI) and neuraminidase-inhibition (NAI) assays. We evaluated how well antibody measured in these specimens represented pre-infection immune status, and measured associations between antibody and influenza vaccination and infection.

RESULTS

Serum specimens were retrieved for 315 participants enrolled during the 2014-2015 season and 339 participants during the 2015-2016 season. Specimens were collected within 3 days of illness onset from 65% of participants. Geometric mean titers (GMTs) did not vary by the number of days from illness onset to specimen collection among influenza positive participants suggesting that measured antibody was representative of pre-infection immune status rather than a de novo response to infection. In both seasons, vaccinated participants had higher HAI and NAI GMTs than unvaccinated. HAI titers against the 2014-2015 A(H3N2) vaccine strain did not correlate with protection from infection with antigenically-drifted A(H3N2) viruses that circulated that season. In contrast, higher HAI titers against the A(H1N1)pdm09 vaccine strain were associated with reduced odds of A(H1N1)pdm09 infection in 2015-2016.

CONCLUSIONS

Serum collected shortly after illness onset at hospital admission can be used to assess correlates of protection against influenza infection. Broader implementation of similar studies would provide an opportunity to understand the successes and shortcomings of current influenza vaccines.

A retrospective analysis of the relationship between race/ethnicity, age at delivery and the risk of gestational diabetes mellitus

Objectives: To investigate the relationship between maternal race/ethnicity, maternal age at delivery, and the risk of gestational diabetes mellitus (GDM).Research design: Patients of Henry Ford Health System who delivered a live singleton child and were diagnosed with or without GDM in 2010-2015 were included. Maternal race/ethnicity, age, body mass index (BMI), parity, GDM in previous pregnancy, smoking status, and insurance membership were collected from the electronic health records. Neighborhood median family income data were obtained from US Census Bureau. Logistic regression analysis was performed to explore the effects of maternal race/ethnicity and age at delivery on the GDM outcome after adjusting for covariates including maternal BMI, parity, previous GDM, smoking status, and neighborhood family income.Results: There were 16,258 women included in the study. Of those, 1801 women (12.5%) were diagnosed with GDM. Delivery at older ages (≥23 years) was associated with a significantly higher risk of a GDM diagnosis than younger ages [<23 years, OR (95% CI) = 2.24 (1.84, 2.73)-5.02 (4.18, 6.03)], however, the risk was not as profound in African American women (OR = 1.65) compared to non-African American women (OR = 2.07). In a multivariable model controlling for age, BMI, parity, previous GDM, smoking status, and the neighborhood family income, the risks of a GDM diagnosis were significantly higher in Asians [OR (95% CI) = 2.81 (2.28, 3.48)], Hispanics [OR (95% CI) = 1.27 (1.05, 1.55)], and Arab Americans [OR (95% CI) = 1.46 (1.20, 1.78)] and lower in African Americans [OR (95% CI) = 0.64 (0.56, 0.74)] as compared to whites.Conclusions: Asians, Hispanics, and Arab Americans have higher risk and African Americans have lower risk of a GDM diagnosis compared to whites. Delivery at an older maternal age increases the risk of GDM diagnosis. Race/ethnicity moderates the association between older maternal age and risk of GDM diagnosis. This study provides information for public health professionals, health practitioners, and pregnant women to be aware of and better understand the risk of GDM as related to race/ethnicity and maternal age.

Multimorbidity is associated with uptake of influenza vaccination

OBJECTIVE

Patients with chronic conditions have higher rates of severe influenza-related illness and mortality. However, influenza vaccination coverage in high-risk populations continues to be suboptimal. We describe the association between cumulative disease morbidity, measured by a previously validated multimorbidity index, and influenza vaccination among community-dwelling adults.

METHODS

We obtained interview and medical record data for participants  ≥18 years who sought outpatient care for influenza-like illness between 2011 and 2016 as part of an outpatient-based study of influenza vaccine effectiveness. We defined cumulative disease morbidity by using medical diagnosis codes to calculate a multimorbidity-weighted index (MWI) for each participant. MWI and influenza vaccination status was evaluated by logistic regression. Akaike information criterion was calculated for all models.

RESULTS

Overall, 1458 (48%) of participants out of a total of 3033 received influenza vaccination. The median MWI was 0.9 (IQR 0.00-3.5) and was higher among vaccinated participants (median 1.6 versus 0.0; p < 0.001). We found a positive linear association between MWI and vaccination, and vaccination percentages were compared between categories of MWI. Compared to patients with no multimorbidity (MWI = 0), odds of vaccination were 17% higher in the second category (MWI 0.01-1.50; [OR: 1.17, 95% CI: 0.92-1.50]), 58% higher in the third category (MWI 1.51-3.00; [OR: 1.58, 95% CI: 1.26-1.99]), 130% higher in the fourth category (MWI 3.01-6.00; [OR: 2.30, 95% CI: 1.78-2.98]) and 214% higher in the fifth category (MWI 6.01-45.00;[OR: 3.14, 95% CI: 2.41-4.10]). Participants defined as high-risk had 86% greater odds of being vaccinated than non-high-risk individuals (OR: 1.86, 95% CI: 1.56-2.21). The AIC was lowest for MWI compared with high-risk conditions.

CONCLUSIONS

Our results suggest a dose response relationship between level of multimorbidity and likelihood of influenza vaccination. Compared with high-risk condition designations, MWI provided improved precision and a better model fit for the measurement of chronic disease and influenza vaccination.

USE OF ELECTRONIC HEALTH RECORDS TO CHARACTERIZE A RARE DISEASE IN THE U.S.: TREATMENT, COMORBIDITIES, AND FOLLOW-UP TRENDS AMONG PATIENTS WITH A CONFIRMED DIAGNOSIS OF ACROMEGALY

OBJECTIVE

Understanding of acromegaly disease management is hampered in the U.S. by the lack of a national registry. We describe medical management in a population with confirmed acromegaly.

METHODS

Inpatient and outpatient electronic health records (EHRs) were used to create a database of de-identified patients assigned the Acromegaly and Gigantism International Classification of Diseases, 9^th revision (ICD-9) code and/or an appropriate pituitary procedure code at 1 of 4 regional hospital systems over a 6- to 11-year period. Information regarding demographics, medical history, labs, procedures, and medications was collected and supplemented with a chart review to validate the diagnosis of acromegaly.

RESULTS

Of 367 patients with validated acromegaly, available records showed that during the years studied, pituitary surgery was performed on 31%, 4% received radiosurgery, and 22% were prescribed a drug indicated for acromegaly. Insulin-like growth factor-1 (IGF-1) levels were measured in 62% of patients, 83% of whom had at least 1 normal value. Coded comorbidities reflect those reported previously in patients with acromegaly, with the exception of esophageal reflux in 20% of patient records. Fewer data regarding acromegaly-specific medications and testing were available for patients aged 65 and older.

CONCLUSION

AcroMEDIC is a U.S. multisite retrospective study of acromegaly that captured medical management in the majority of patients included in the cohort. Chart review highlighted the importance of verification of coded diagnoses. Most of the acromegaly-related comorbidities identified here are known to increase with age and obesity. Patients ≥65 appeared to have less active management/monitoring of their disease. Medical attention should be directed to this population to address evolving needs over time.

ABBREVIATIONS

AcroMEDIC = Acromegaly Multisite Electronic Data Innovative Consortium; BMI = body mass index; CCI = Charlson Comorbidity Index; EHR = electronic health record; GH = growth hormone; GHRA = growth hormone receptor antagonist; ICD-9 = International Classification of Diseases, 9^th revision; IGF-1 = insulin-like growth factor-1; SSA = somatostatin analogue.

Complementary and alternative medicine use and hormonal therapy initiation in women with hormone receptor-positive breast cancer: The BQUAL study

e13097

Background: Use of complementary and alternative medicine (CAM) has been associated with initiation of breast cancer chemotherapy.We examined the association of CAM use with initiation of hormonal therapy in women with hormone receptor-positive (HR+) breast cancer. Methods: In aprospective cohort study designed to examine predictors of breast cancer treatment initiation and adherence, 699 women with non-metastatic stage I-III breast cancer and aged < 70 years were recruited from Kaiser Permanente Northern California during 2006-2010. Baseline interviews assessed current use of five CAM modalities (dietary supplements: vitamins/minerals, herbs/botanicals, other natural products; mind/body therapies: mind-body self-practice, mind-body practitioner-based). Based on electronic pharmacy records, hormonal therapy initiation was defined as ≥2 prescriptions of ≥30 pills of any hormonal therapy drug within 1 year after diagnosis. Multivariable logistic regression analysis examined a priori hypotheses testing whether current CAM use was associated with hormonal therapy initiation, adjusted for demographic, tumor characteristics, and cancer treatment. Results: Of the 699 participants,552 were diagnosed with HR+ breast cancer. Among the 552 women with HR+ breast cancer, 494 (89%) used at least 1 modality of CAM, 414 (76%) used dietary supplements, and 391 (71%) used mind/body therapies. Within 1 year after diagnosis, 481 (87%) participants initiated hormonal therapy. The rate of initiation was comparable between CAM users and non-users (87% vs. 86%). In multivariable analyses, baseline overall CAM use was not associated with initiation of hormonal therapy. However, vitamins/minerals users were more likely to initiate hormonal therapy compared to non-users (OR = 1.87, 95% CI: 1.04, 3.51). Conclusions: CAM use was high among early stage HR+ breast cancer patients. Overall, use of CAM was not associated with hormonal therapy initiation. Current use of vitamins/minerals was associated with higher initiation rate of hormonal chemotherapy.

Hepatitis C treatment failure is associated with increased risk of hepatocellular carcinoma

Sustained virological response (SVR) to antiviral therapy for hepatitis C (HCV) reduces risk of hepatocellular carcinoma (HCC), but there is little information regarding how treatment failure (TF) compares to lack of treatment. We evaluated the impact of treatment status on risk of HCC using data from the Chronic Hepatitis Cohort Study (CHeCS-an observational study based in four large US health systems, with up to 7 years of follow-up on patients). Multivariable analyses were used to adjust for bias in treatment selection, as well as other covariates, followed by sensitivity analyses. Among 10 091 HCV patients, 3681 (36%) received treatment, 2099 (57%) experienced treatment failure (TF), and 1582 (43%) of these achieved sustained virological response (SVR). TF patients demonstrated almost twice the risk of HCC than untreated patients [adjusted hazard ratio (aHR) = 1.95, 95% confidence interval (CI) 1.50-2.53]; this risk persisted across all stages of fibrosis. Several sensitivity analyses validated these results. Although African Americans were at increased risk of treatment failure, they were at lower risk for HCC and all-cause mortality compared to White patients. SVR patients had lower risk of HCC than TF patients (aHR = 0.48, CI 0.31-0.73), whereas treatment - regardless of outcome - reduced all-cause mortality (aHR = 0.45, CI 0.34-0.60 for SVR patients; aHR = 0.78, CI 0.65-0.93 for TF patients).

Development and Validation of Algorithms to Identify Statin Intolerance in a US Administrative Database

OBJECTIVES

To develop and validate algorithms to define statin intolerance (SI) in an administrative database using electronic medical records (EMRs) as the reference comparison.

METHODS

One thousand adults with one or more qualifying changes in statin therapy and one or more previous diagnoses of hyperlipidemia, hypercholesterolemia, or mixed dyslipidemia were identified from the Henry Ford Health System administrative database. Data regarding statin utilization, comorbidities, and adverse effects were extracted from the administrative database and corresponding EMR. Patients were stratified by cardiovascular (CV) risk. SI was classified as absolute intolerance or titration intolerance on the basis of changes in statin utilization and/or the occurrence of adverse effects and laboratory testing for creatine kinase. Measures of concordance (Cohen's kappa [κ]) and accuracy (sensitivity, specificity, positive predictive value [PPV], and negative predictive value) were calculated for the administrative database algorithms.

RESULTS

Half of the sample population was white, 52.9% were women, mean age was 60.6 years, and 35.7% were at high CV risk. SI was identified in 11.5% and 14.0%, absolute intolerance in 2.2% and 3.1%, and titration intolerance in 9.7% and 11.8% of the patients in the EMR and the administrative database, respectively. The algorithm identifying any SI had substantial concordance (κ = 0.66) and good sensitivity (78.1%), but modest PPV (64.0%). The titration intolerance algorithm performed better (κ = 0.74; sensitivity 85.4%; PPV 70.1%) than the absolute intolerance algorithm (κ = 0.40; sensitivity 50%; PPV 35.5%) and performed best in the high CV-risk group (n = 353), with robust concordance (κ = 0.73) and good sensitivity (80.9%) and PPV (75.3%).

CONCLUSIONS

Conservative but comprehensive algorithms are available to identify SI in administrative databases for application in real-world research. These are the first validated algorithms for use in administrative databases available to decision makers.

The occurrence of hyponatremia and its importance as a prognostic factor in a cross-section of cancer patients

Background

Hyponatremia is prognostic of higher mortality in some cancers but has not been well studied in others. We used a longitudinal design to determine the incidence and prognostic importance of euvolemic and hypervolemic hyponatremia in patients following diagnosis with lymphoma, breast (BC), colorectal (CRC), small cell lung (SCLC), or non-small cell lung cancer (NSCLC).

Methods

Medical record and tumor registry data from two large integrated delivery networks were combined for patients diagnosed with lymphoma, BC, CRC, or lung cancers (2002–2010) who had ≥1 administration of radiation/chemotherapy within 6 months of diagnosis and no evidence of hypovolemic hyponatremia. Hyponatremia incidence was measured per 1000 person-years (PY). Cox proportional hazard models assessed the prognostic value of hyponatremia as a time-varying covariate on overall survival (OS) and progression-free survival (PFS).

Results

Hyponatremia incidence (%, rate) was 76 % each, 1193 and 2311 per 1000 PY, among NSCLC and SCLC patients, respectively; 37 %, 169 in BC; 64 %, 637 in CRC, and 60 %, 395 in lymphoma. Hyponatremia was negatively associated with OS in BC (HR 3.7; P = <.01), CRC (HR 2.4; P < .01), lung cancer (HR 2.4; P < .01), and lymphoma (HR 4.5; P < .01). Hyponatremia was marginally associated with shorter PFS (HR 1.3, P = .07) across cancer types.

Conclusions

The incidence of hyponatremia is higher than previously reported in lung cancer, is high in lymphoma, BC, and CRC and is a negative prognostic indicator for survival. Hyponatremia incidence in malignancy may be underestimated. The effects of hyponatremia correction on survival in cancer patients require further study.

Validation of Self-Report of Chest X-Ray Exam at a Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial Center

It is imperative to measure the degree of contamination throughout the course of randomized controlled trials, as contamination, the receipt of the intervention arm regimen by control arm participants, can affect trial power. In the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial, contamination was estimated through use of the self-administered Health Status Questionnaire (HSQ) annually to a randomly-selected subset of control arm participants. We examined agreement of self-report of chest x-ray on the HSQ with clinic records at one of the 10 PLCO screening centers (Henry Ford Health System, or HFHS). We focus on HFHS participants covered by the Health Alliance Plan (HAP), a managed health care insurance plan owned and operated by HFHS, because claims for care received both at HFHS and other facilities are available in HFHS databases for HAP enrollees. We examined agreement for the six years prior to HSQ completion, with HFHS clinic records considered to be the gold standard. For those who had complete HAP coverage during the six years, percent agreement was 0.69, sensitivity was 0.84, and positive predictive value was 0.76. Specificity and negative predicted value were low, however (0.28 and 0.38, respectively), and Cohen's kappa was 0.13. For groups with incomplete or no HAP coverage, and when timing of exam was considered, performance measures typically became lower, in some instances below 0.20. These data suggest that self-report of chest x-ray screening may not be accurate, although high prevalence of chest x-ray may make performance measures less interpretable.

Influenza Vaccine Effectiveness Against Antigenically Drifted Influenza Higher Than Expected in Hospitalized Adults: 2014-2015

BACKGROUND

The 2014-2015 influenza season was severe, with circulating influenza A (H3N2) viruses that were antigenically drifted from the vaccine virus. Reported vaccine effectiveness (VE) estimates from ambulatory care settings were markedly decreased.

METHODS

Adults, hospitalized at 2 hospitals in southeast Michigan for acute respiratory illnesses, defined by admission diagnoses, of ≤10 days duration were prospectively enrolled. Throat and nasal swab specimens were collected, combined, and tested for influenza by real-time reverse transcription polymerase chain reaction. VE was estimated by comparing the vaccination status of those testing positive for influenza with those testing negative in logistic regression models adjusted for age, sex, hospital, calendar time, time from illness onset to specimen collection, frailty score, and Charlson comorbidity index (CCI).

RESULTS

Among 624 patients included in the analysis, 421 (68%) were vaccinated, 337 (54%) were female, 220 (35%) were age ≥65 years, and 92% had CCI > 0, indicating ≥1 comorbid conditions. Ninety-eight (16%) patients tested positive for influenza A (H3N2); among 60 (61%) A (H3N2) viruses tested by pyrosequencing, 53 (88%) belonged to the drifted 3C.2a genetic group. Adjusted VE was 43% (95% confidence interval [CI], 4-67) against influenza A (H3N2); 40% (95% CI, -13 to 68) for those <65 years, and 48% (95% CI, -33 to 80) for those ≥65 years. Sensitivity analyses largely supported these estimates.

CONCLUSIONS

VE estimates appeared higher than reports from similar studies in ambulatory care settings, suggesting that the 2014-2015 vaccine may have been more effective in preventing severe illness requiring hospitalization.

Abstract P5-08-49: Incidence and prognostic importance of hyponatremia in a cohort of patients with breast cancer

Background: It has been suggested that hyponatremia (HN) may be a negative prognostic factor in patients with cancer but little research has been conducted specifically in breast cancer (BC). We measured the incidence of hyponatremia (hypervolemic and euvolemic) after BC diagnosis and its prognostic importance for progression free (PFS) and overall survival (OS).

Methods: This retrospective cohort analysis utilized data from the Henry Ford Health System electronic medical record, tumor registry, and administrative databases. Study data were collected electronically and via medical record abstraction. Adults diagnosed (2002-2010) with incident invasive BC were selected if they had a known disease stage at the time of tumor registration, were classified as an analytic case, had ≥ 1 administration of chemo/radiation therapy ≤ 6 months from diagnosis, met continuous enrollment thresholds, and did not experience hypovolemic HN post index. Only the first tumor registered from each patient was considered study-eligible. Hypervolemic or euvolemic HN incidence (serum sodium ≤ 135 mEq/L) was measured per 1000 person-years (PY) of observation and classified as mild (131–135 mEq/L), moderate (125–130 mEq/L) or severe (&lt;125 mEq/L) based on the lowest observed value. A Cox proportional hazards model was used to assess the prognostic value of HN as a time-varying covariate on PFS and OS while controlling for age, race, income, morphology code, diagnosis year, cancer stage at diagnosis, performance status at diagnosis, and hormone receptor status.

Results: 527 patients were eligible (mean [SD] age 56.4±11.3 years, 61% Caucasian). Mean (SD) follow-up was 3.7±2.8 years. Eighty-five percent of patients had infiltrating ductal carcinoma; 72% and 65% had estrogen or progesterone sensitive tumor, respectively; 35% were HER2 positive; and 15% had triple negative disease. Eighty-two percent of patients had early stage (I, II) disease at time of diagnosis. HN episodes (n=377) occurred in 204 patients (39%) at a rate of 193 per 1000 PY (95% CI, 174–213.5), with 89% of the total episodes (337/377) classified as mild, 10% (36/377) as moderate, and 1% (4/377) as severe. Additionally, 7% of all BC patients (37/527) had at least one episode of moderate/severe HN. Median time to first HN episode was 174.5 days and the median HN episode duration was 24.0 days. Five year OS in patients developing HN was 92%, compared to 97% in patients who never developed HN. Hazard ratio (95% CI, p-value) for OS in the HN group was 4.4 (1.5-12.7; p=0.006) after controlling for age, diagnosis year, race, income, morphology, cancer stage, performance status, and hormone receptor status. Fifty patients had progressive disease during follow-up with a mean (SD) time to progression of 763.4 (758.1) days. Hazard ratio (95% CI, p-value) for PFS in the HN group was 1.4 (0.8-2.7; p=0.262) after controlling for age, race, income, morphology, cancer stage, performance status, and hormone receptor status.

Conclusions: Incidence of hypervolemic or euvolemic HN is high (39%) after a BC diagnosis, and the occurrence is associated with significantly poorer OS. A significant impact on disease progression was not observed.

Citation Format: Castillo JJ, Glezerman IG, Boklage SH, Lamerato LE, Chiodo JA, Tidwell BA, Schulman KL. Incidence and prognostic importance of hyponatremia in a cohort of patients with breast cancer. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P5-08-49.

Abstract PD4-04: Psychosocial factors related to interruptions in adjuvant hormonal therapy among women with breast cancer: The breast cancer quality of care study (BQUAL)

Background. Adjuvant hormonal therapy (HT) for hormone-sensitive breast cancer decreases risk of breast cancer recurrence and improves survival. However, some women are non-adherent to this life-saving treatment.

Methods. In a cohort of women recruited at diagnosis of breast cancer in an integrated healthcare system, we investigated factors related to HT interruption (≥90 day gap). Serial interviews were conducted at baseline and during treatment to examine psychological factors as well as sociodemographic factors, tumor characteristics, and treatment factors. A series of multivariate models assessed potential predictors of HT interruptions.

Results. Of the 569 women in our cohort who initiated HT, 137 (24%) interrupted it, including 18 (3%) who did so prior to the first follow-up interview. In a multivariate analysis of clinical and demographic factors, only household income remained associated with HT interruption (OR 0.42, 95%CI 0.24-0.76). At first follow-up, after controlling for income, race and age, lower scores on all quality of life subscales, lower scores on global treatment satisfaction, and poorer scores on the intrusive and avoidant thought subscales of the Impact of Events scale were associated with higher odds of HT interruptions (P&lt;0.001 for all predictors). Scores on social support and on interpersonal processes of care measures were not associated with HT interruptions. However, a higher score on the single question "How often did your doctor speak too fast?" was associated with higher risk of HT interruptions (OR 1.32, p=0.02).

Conclusions: Patients under greater duress and those with lower physical, functional, emotional or social quality of life appeared to be at the highest risk of HT interruption and thus received poorer quality care. A better understanding of psychological factors that can result in poor quality care may pave the way to targeted interventions to improve adherence.

Citation Format: Hershman DL, Kushi LH, Hillyer GC, Coromilis E, Buono D, Lamerato LE, Bovbjerg DH, Mandelblatt8 JS, Tsai W-Y, Jacobson JS, Wright JD, Neugut AI. Psychosocial factors related to interruptions in adjuvant hormonal therapy among women with breast cancer: The breast cancer quality of care study (BQUAL). [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr PD4-04.

Cost Burden of Chronic Pain Patients in a Large Integrated Delivery System in the United States

OBJECTIVES

To estimate all-cause healthcare resource utilization and costs among chronic pain patients within an integrated healthcare delivery system in the United States.

METHODS

Electronic medical records and health claims data from the Henry Ford Health System were used to determine healthcare resource utilization and costs for patients with 24 chronic pain conditions. Patients were identified by ≥ 2 ICD-9-CM codes ≥ 30 days apart from January to December, 2010; the first ICD-9 code was the index event. Continuous coverage for 12 months pre- and postindex was required. All-cause direct medical costs were determined from billing data.

RESULTS

A total of 12,165 patients were identified for the analysis. After pharmacy, the most used resource was outpatient visits, with a mean of 18.8 (SD 13.2) visits per patient for the postindex period; specialty visits accounted for 59.0% of outpatient visits. Imaging was utilized with a mean of 5.2 (SD 5.5) discrete tests per patient, and opioids were the most commonly prescribed medication (38.7%). Annual direct total costs for all conditions were $386 million ($31,692 per patient; a 40% increase from the pre-index). Pharmacy costs comprised 14.3% of total costs, and outpatient visits were the primary cost driver.

CONCLUSIONS

Chronic pain conditions impose a substantial burden on the healthcare system, with musculoskeletal conditions associated with the highest overall costs. Costs appeared to be primarily related to use of outpatient services. This type of research supports integrated delivery systems as a source for assessing opportunities to improve patient outcomes and lower the costs for chronic pain patients.

Lessons in Medical Record Abstraction from the Prostate, Lung, Colorectal, and Ovarian (PLCO) National Screening Trial

The most rigorous and accurate approach to evaluating clinical events in cancer screening studies is to use data obtained through medical record abstraction (MRA). Although MRA is complex, the particulars of the procedure-such as the specific training and quality assurance processes, challenges of implementation, and other factors that influence the quality of abstraction--are usually not described in reports of studies that employed the technique. In this paper, we present the details of MRA activities used in the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial, which used MRA to determine primary and secondary outcomes and collect data on other clinical events. We describe triggers of the MRA cycle and the specific tasks that were part of the abstraction process. We also discuss training and certification of abstracting staff, and technical methods and communication procedures used for data quality assurance. We include discussion of challenges faced and lessons learned.

Prevalence of cirrhosis in hepatitis C patients in the Chronic Hepatitis Cohort Study (CHeCS): a retrospective and prospective observational study

OBJECTIVES

The severity of liver disease in the hepatitis C virus (HCV)-infected population in the United States remains uncertain. We estimated the prevalence of cirrhosis in adults with chronic hepatitis C (CHC) using multiple parameters including liver biopsy, diagnosis/procedure codes, and a biomarker.

METHODS

Patients enrolled in the Chronic Hepatitis Cohort Study (CHeCS) who received health services during 2006-2010 were included. Cirrhosis was identified through liver biopsy reports, diagnosis/procedure codes for cirrhosis or hepatic decompensation, and Fibrosis-4 (FIB-4) scores ≥5.88. Demographic and clinical characteristics associated with cirrhosis were identified through multivariable logistic modeling.

RESULTS

Among 9,783 patients, 2,788 (28.5%) were cirrhotic by at least one method. Biopsy identified cirrhosis in only 661 (7%) patients, whereas FIB-4 scores and diagnosis/procedure codes for cirrhosis and hepatic decompensation identified cirrhosis in 2,194 (22%), 557 (6%), and 482 (5%) patients, respectively. Among 661 patients with biopsy-confirmed cirrhosis, only 356 (54%) had an International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) code for cirrhosis. Older age, male gender, Asian race, Hispanic ethnicity, genotype 3 infection, HIV coinfection, diabetes, history of antiviral therapy, and history of alcohol abuse were independently associated with higher odds of cirrhosis (all, P<0.05). Conversely, private health insurance coverage, black race, and HCV genotype 2 were associated with lower odds of cirrhosis.

CONCLUSIONS

A high proportion of patients with biopsy-confirmed cirrhosis are not assigned ICD-9 codes for cirrhosis. Consequently, ICD-9 codes may not be reliable as the sole indicator of the prevalence of cirrhosis in cohort studies. Use of additional parameters suggests a fourfold higher prevalence of cirrhosis than is revealed by biopsy alone. These findings suggest that cirrhosis in CHC patients may be significantly underdocumented and underdiagnosed.

Racial differences in breast cancer survival in a large urban integrated health system

BACKGROUND

African American (AA) women are known to have poorer breast cancer survival than whites, and the differences may be related to underlying disparities in their clinical presentation or access to care. This study evaluated the relationship between demographic, treatment, and socioeconomic factors and breast cancer survival among women in southeast Michigan.

METHODS

The population included 2387 women (34% AA) with American Joint Committee on Cancer stage I to III breast cancer who were treated at the Henry Ford Health System (HFHS) from 1996 through 2005. Linked data sets from the HFHS, the Metropolitan Detroit Cancer Surveillance System, and the US Census Bureau were used to obtain demographic and clinical information. Comorbidities were classified with the modified Charlson comorbidity index (CCI). Economic deprivation was categorized with a census tract-based deprivation index (DI), which was stratified into 5 quintiles of increasing socioeconomic disadvantage.

RESULTS

Compared with whites, AA women were significantly more likely to have larger, hormone receptor-negative tumors and more comorbidities and to reside in an economically deprived area. In an unadjusted analysis, AAs had a significantly higher risk of death (hazard ratio [HR], 1.36; 95% confidence interval [CI], 1.16-1.59); however, after adjustments for clinical (age, stage, hormone receptor, and CCI) and societal factors (DI), the effect of race was not significant (HR, 1.13 [95% CI, 0.96-1.34] , and HR, 0.97 [0.80-1.19] respectively).

CONCLUSIONS

Racial differences in breast cancer survival can be explained by clinical and socioeconomic factors. Nonetheless, AA women with breast cancer remain disproportionately affected by unfavorable tumor characteristics and economic deprivation, which likely contribute to their increased overall mortality.

Abstract 280: Defining Statin Intolerance Among High Cardiovascular Risk Patients: Do US Administrative Databases Lend Themselves to Identification Of Statin Intolerance?

Introduction: The clinical and economic impact of statin intolerance (SI) in high CV risk patients is unknown due, in part, to a lack of consensus in its definition. We sought to define and validate an SI algorithm for use in an administrative database (AD) among high-CV risk patients

Methods: Adults with ≥1 qualifying change (See Table 1) in statin therapy and ≥1 prior diagnosis of hyperlipidemia, hypercholesterolemia, or mixed dyslipidemia were identified from the AD of the Health Alliance Plan at Henry Ford Health System (HFHS). A sample of 1000 patients was drawn from the pool of eligible adults and stratified by high CV risk based on presence of comorbid conditions including diabetes, coronary heart disease, and peripheral artery disease. Statin utilization and adverse events data were abstracted both from the AD and the HFHS electronic medical record (EMR). SI was defined using both a primary definition inclusive of all possible statin related adverse events and a secondary definition that included only musculoskeletal events. SI was categorized as absolute (AI) or titration (TI) intolerance. The performance of the AD algorithm was assessed using measures of concordance (Cohen’s kappa [κ]) and accuracy (sensitivity, specificity, positive predictive value [PPV]) with the EMR as reference.

Results: A total of 353 patients (48% female, 44% Caucasian, mean (SD) age 63 (12) years) were identified as high CV risk with 33% having a history of CHD, 77% diabetes and 2% PAD. Forty-two percent of patients were on simvastatin, 35% atorvastatin, 11% lovastatin, 7% rosuvastatin and 6% pravastatin/fluvastatin. Table 1 characterizes the validation sample. SI was identified in 19.3% and 20.7%, AI in 3.1% and 2.8%, and TI in 16.7% and 18.7% of patients in the EMR and AD, respectively. The algorithm identifying any SI had robust concordance (κ=0.73), good sensitivity (80.9%) and PPV (75.3%). The TI algorithm performed better (κ=0.78, sensitivity=86.4%, PPV=77.3%) than the AI algorithm (κ=0.56, sensitivity=54.5%, PPV=60.0%). Specificity was high (>94%) across all 3 algorithms.

Conclusion: This study successfully defined SI among high-CV risk patients using an evidence-based validated algorithm. To our knowledge, this is the first such algorithm for use in AD to be made available to decision-makers.

Comparative effectiveness research of chronic hepatitis B and C cohort study (CHeCS): improving data collection and cohort identification

BACKGROUND AND AIMS

The Chronic Hepatitis Cohort Study (CHeCS) is a longitudinal observational study of risks and benefits of treatments and care in patients with chronic hepatitis B (HBV) and C (HCV) infection from four US health systems. We hypothesized that comparative effectiveness methods-including a centralized data management system and an adaptive approach for cohort selection-would improve cohort selection while controlling data quality and reducing the cost.

METHODS

Cohort selection and data collection were performed primarily via the electronic health record (EHR); cases were confirmed via chart abstraction. Two parallel sources fed data to a centralized data management system: direct EHR data collection with common data elements, and chart abstraction via electronic data capture. An adaptive Classification and Regression Tree (CART) identified a set of electronic variables to improve case ascertainment accuracy.

RESULTS

Over 16 million patient records were collected on 23 case report forms in 2006-2008. The vast majority of data (99.2%) were collected electronically from EHR; only 0.8% was collected via chart abstraction. Initial electronic criteria identified 12,144 chronic hepatitis patients; 10,098 were confirmed via chart abstraction with positive predictive values (PPV) 79 and 83% for HBV and HCV, respectively. CART-optimized models significantly increased PPV to 88 for HBV and 95% for HCV.

CONCLUSIONS

CHeCS is a comparative effectiveness research project that leverages electronic centralized data collection and adaptive cohort identification approaches to enhance study efficiency. The adaptive CART model significantly improved the positive predictive value of cohort identification methods.

Non-compliance with the initial screening exam visit in the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial

OBJECTIVE

Identify predictors of non-compliance with first round screening exams in the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial.

METHOD

The PLCO was conducted from 1993 to 2011 at 10 US institutions. A total of 154,897 healthy men and women ages 55-74 years were randomized. Intervention arm participants were invited to receive gender-appropriate screening exams for prostate, lung, colorectal and ovarian cancer. Using intervention-arm data (73,036 participants), non-compliance percentages for 13 covariates were calculated, as were unadjusted and adjusted odds ratios (ORs), and 95% confidence intervals. Covariates included demographic factors as well as factors specific to PLCO (e.g., method of consent, distance from screening center).

RESULTS

The rate of non-compliance was 11% overall but varied by screening center. Significant associations were observed for most covariates but indicated modest increases or decreases in odds. An exception was the use of a two-step consent process (consented intervention arm participants for exams after randomization) relative to a one-step process (consented all participants prior to randomization) (OR: 2.2, 95% CI: 2.0-2.5). Non-compliance percentages increased with further distance from screening centers, but ORs were not significantly different from 1.

CONCLUSIONS

Many factors modestly influenced compliance. Consent process was the strongest predictor of compliance.

Geospatial relationships of air pollution and acute asthma events across the Detroit-Windsor international border: study design and preliminary results

The Geospatial Determinants of Health Outcomes Consortium (GeoDHOC) study investigated ambient air quality across the international border between Detroit, Michigan, USA and Windsor, Ontario, Canada and its association with acute asthma events in 5- to 89-year-old residents of these cities. NO2, SO2, and volatile organic compounds (VOCs) were measured at 100 sites, and particulate matter (PM) and polycyclic aromatic hydrocarbons (PAHs) at 50 sites during two 2-week sampling periods in 2008 and 2009. Acute asthma event rates across neighborhoods in each city were calculated using emergency room visits and hospitalizations and standardized to the overall age and gender distribution of the population in the two cities combined. Results demonstrate that intra-urban air quality variations are related to adverse respiratory events in both cities. Annual 2008 asthma rates exhibited statistically significant positive correlations with total VOCs and total benzene, toluene, ethylbenzene and xylene (BTEX) at 5-digit zip code scale spatial resolution in Detroit. In Windsor, NO2, VOCs, and PM10 concentrations correlated positively with 2008 asthma rates at a similar 3-digit postal forward sortation area scale. The study is limited by its coarse temporal resolution (comparing relatively short term air quality measurements to annual asthma health data) and interpretation of findings is complicated by contrasts in population demographics and health-care delivery systems in Detroit and Windsor.

Noninvasive serum fibrosis markers for screening and staging chronic hepatitis C virus patients in a large US cohort

BACKGROUND

Liver biopsy remains critical for staging liver disease in hepatitis C virus (HCV)-infected persons, but is a bottleneck to evaluation, follow-up, and treatment of HCV. Our analysis sought to validate APRI (aspartate aminotransferase [AST]-to-platelet ratio index) and FIB-4, an index from serum fibrosis markers (alanine aminotransferase [ALT], AST, and platelets plus patient age) to stage liver disease.

METHODS

Biopsy results from HCV patients in the Chronic Hepatitis Cohort Study were mapped to an F0-F4 equivalent scale; APRI and FIB-4 scores at the time of biopsy were then mapped to the same scale.

RESULTS

We identified 2372 liver biopsies from HCV-infected patients with contemporaneous laboratory values for imputing APRI and FIB-4. Fibrosis stage distributions by the equivalent biopsy scale were 267 (11%) F0; 555 (23%) F1; 648 (27%) F2; 394 (17%) F3; and 508 (21%) F4. Mean APRI and FIB-4 values significantly increased with successive fibrosis levels (P < .05). The areas under the receiver operating characteristic curve (AUROC) analysis distinguishing severe (F3-F4) from mild-to-moderate fibrosis (F0-F2) were 0.80 (95% confidence interval [CI], .78-.82) for APRI and 0.83 (95% CI, .81-.85) for FIB-4. There was a significant difference between the AUROCs of FIB-4 and APRI (P < .001); 88% of persons who had a FIB-4 score ≥2.0 were at stage F2 or higher.

CONCLUSIONS

In a large observational cohort, FIB-4 was good at differentiating 5 stages of chronic HCV infection. It can be useful in screening patients who need biopsy and therapy, for monitoring patients with less advanced disease, and for longitudinal studies.

Assessment of multiple health risks in a single obstructive sleep apnea population

STUDY OBJECTIVES

In order to provide a comprehensive estimate of the health risks for OSA patients, we analyzed multiple outcomes and independent predictors of these outcomes in an OSA population evaluated and followed at one sleep center.

METHODS

Cox proportional hazard regression analyses were used in an 8-year follow-up analysis of consecutive OSA patients (N = 1025) and non-apneic snorers (apnea-hypopnea index < 5, N = 494).

RESULTS

In our fully adjusted model, independent variables predictive of all-cause mortality, myocardial infarction, cerebral vascular accident, and pulmonary embolus were: older age, male gender, and history of cardiovascular diseases or procedures. In examining subgroups based on age and gender, severe OSA (AHI ≥ 30) was one of the independent predictors of mortality in males and in patients < 50 years old. Severe OSA interacted with maleness, age, and hypertension to predict mortality and myocardial infarction. CPAP use ≥ 4 h/night was associated with lower mortality rates in males and those ≥ 50 years old with severe OSA.

CONCLUSIONS

Mortality and cardiovascular event outcomes were predicted by demographics and cardiovascular disease history more commonly than by OSA severity. OSA severity was an important predictor of mortality in male and young OSA patients. CPAP use appeared protective in older and male severe OSA patients.

Benign prostatic hyperplasia: racial differences in treatment patterns and prostate cancer prevalence

OBJECTIVE

• To compare prostate cancer, prostate-related surgery and acute urinary retention rates, as well as associated healthcare resource use over 11 years in African American and Caucasian men with benign prostatic hyperplasia (BPH).

PATIENTS AND METHODS

• The BPH-related medical and surgical charges and events were determined for 398 African American men and 1656 Caucasian men followed for a mean of 10.2 years within a health maintenance organization. • Racial differences in clinical outcomes were evaluated using time-to-event analysis, stratifying results by baseline prostate-specific antigen (PSA) values.

RESULTS

• Risk of a prostate cancer diagnosis was 2.2 times greater in African American than Caucasian men (95% CI 1.48-3.35, P < 0.001) in analyses adjusting for serum PSA level. • Although African Americans were more likely to receive medical therapy for symptoms of BPH than Caucasians (43.5% vs 37.2%, respectively; P= 0.029), there were no clinically meaningful differences with respect to subsequent acute urinary retention or BPH-related surgery between them, or BPH-related medical charges (US $407 vs US $405 per month).

CONCLUSION

• As evidenced by this analysis of 'real-world' clinical practice, African Americans with BPH have a much greater risk of developing prostate cancer than similar Caucasian men highlighting the need for education and early detection in this population.

Prevalence of stroke/transient ischemic attack among patients with acute coronary syndromes in a real-world setting

BACKGROUND

Atherothrombosis is a systemic disease that may manifest as acute ischemic events in multiple vascular beds. Patients who have experienced an atherothrombosis-related ischemic event in 1 vascular bed are at risk for developing ischemic events in other vascular beds. Antiplatelet therapy demands an understanding of the balance between arterial thrombosis benefit and adverse event risk. Clinical trials indicate that dual antiplatelet therapy with aspirin and the newer thienopyridines increases the risk of bleeding in patients with acute coronary syndromes (ACS) with prior cerebrovascular events. Informed clinical decision making requires a better understanding of the real-world prevalence of cerebrovascular events.

OBJECTIVE AND PURPOSE

To estimate the prevalence of stroke and/or transient ischemic attack (TIA) among patients with ACS within US health plan populations.

METHODS

A retrospective, observational cohort study was conducted of patients with ACS in 5 health care claims databases. The index event was defined as the first documented inpatient health care claim for myocardial infarction or unstable angina. Patients with ≥12 months of pre-index medical care encounter information were included. Stroke/TIA was identified by the first health care claim for these conditions any time prior to or within 90 days following the index ACS event.

RESULTS

Across all databases, between 3.8% and 15.7% of patients with ACS had prior stroke/TIA and between 3.4% and 11.7% of patients with ACS with no history of cerebrovascular events had documented stroke/TIA following the index ACS hospitalization.

CONCLUSION

Despite important differences between the various database populations, there is a high prevalence of documented stroke/TIA in patients with ACS both prior to and following the ACS event. These real-world findings, set within the context of the increased bleeding risk observed with the newer thienopyridines, are important considerations when selecting antiplatelet therapy for patients with ACS.

Race and colorectal cancer disparities: health-care utilization vs different cancer susceptibilities

BACKGROUND

It is unclear whether the disproportionately higher incidence and mortality from colorectal cancer among blacks compared with whites reflect differences in health-care utilization or colorectal cancer susceptibility.

METHODS

A total of 60, 572 non-Hispanic white and black participants in the ongoing Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial underwent trial-sponsored screening flexible sigmoidoscopy (FSG) without biopsy at baseline in 10 geographically dispersed centers from November 1993 to July 2001. Subjects with polyps or mass lesions detected by FSG were referred to their physicians for diagnostic workup, the cost of which was not covered by PLCO. The records of follow-up evaluations were collected and reviewed. We used log binomial modeling with adjustment for age, education, sex, body mass index, smoking, family history of colorectal cancer, colon examination within previous 3 years, personal history of polyps, and screening center to examine whether utilization of diagnostic colonoscopy and yield of neoplasia differed by race.

RESULTS

Among 57 561 whites and 3011 blacks who underwent FSG, 13,743 (23.9%) and 767 (25.5%) had abnormal examinations, respectively. A total of 9944 (72.4%) whites and 480 (62.6%) blacks had diagnostic colonoscopy within 1 year following the abnormal FSG screening. When compared with whites, blacks were less likely to undergo diagnostic evaluation (adjusted risk ratio = 0.88, 95% confidence interval = 0.83 to 0.93). Overall, among subjects with diagnostic colonoscopy (n = 10 424), there was no statistically significant difference by race in the prevalence of adenoma, advanced adenoma, advanced pathology in small adenomas (high-grade dysplasia or villous histology in adenomas <10 mm), or colorectal cancer.

CONCLUSIONS

We observed a lower follow-up for screen-detected abnormalities among blacks when compared with whites but little difference in the yield of colorectal neoplasia. Health-care utilization may be playing more of a role in colorectal cancer racial disparity than biology.

Recruitment in the prostate, lung, colorectal, and ovarian (PLCO) cancer screening trial: the first phase of recruitment at Henry Ford Health System

OBJECTIVE

Recruitment of healthy subjects to long-term randomized controlled trials (RCTs) of cancer prevention or early detection has proven to be a difficult task. To quantify recruitment yield as well as characteristics of successfully recruited participants, we examined recruitment outcomes at 1 of the 10 centers participating in the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial, a National Cancer Institute-funded RCT of cancer screening modalities.

MATERIALS AND METHODS

During the early recruitment phase of PLCO (1993-1997), data on recruitment outcome were collected at the Henry Ford Health System (HFHS) in Detroit, Michigan. In this phase, HFHS identified potential participants using patient databases. Records were used to assess recruitment success by age, sex, race, household income (using area-based U.S. Census data), and preexisting morbidity. Logistic regression was used to assess whether enrollment success differed significantly according to these factors.

RESULTS

Of 74,139 persons ages 55 to 74 invited by HFHS to participate, 8,250 (11%) ;enrolled. In multivariate analyses, the odds of enrolling were modestly but significantly higher for women, Caucasians, persons in their 60's, and persons living in census blocks with higher median household income. Persons with two or more preexisting morbidities had significantly lower odds of enrolling compared to those with one or no preexisting morbidities.

CONCLUSIONS

These data suggest that only a small fraction of persons invited to enroll in long-term RCTs of cancer screening modalities actually do so. In this urban, Midwestern setting, certain characteristics including age, race, and income influenced recruitment success, albeit modestly.

Diffusion of aromatase inhibitors for breast cancer therapy between 1996 and 2003 in the Cancer Research Network

PURPOSE

Clinical trials demonstrated adjuvant aromatase inhibitor treatment is superior for decreasing breast cancer recurrence risk over adjuvant tamoxifen treatment as early as 2001. Yet clinical use for adjuvant treatment was not recommended by the American Society of Clinical Oncology until 2004. Aromatase inhibitor uptake after the first public presentation of randomized trial results but before the release of national guidelines is unclear. We evaluated diffusion of aromatase inhibitor dispensings for breast cancer treatment in integrated healthcare delivery systems across the United States.

METHODS

We collected automated data for 13,245 women enrolled at seven integrated healthcare delivery systems in the Cancer Research Network. All women were aged >55 and diagnosed with estrogen receptor positive, invasive breast cancer between 1996 and 2003. We used electronic pharmacy data to identify aromatase inhibitor and tamoxifen dispensings through 2004. We evaluated the proportions of women who received hormone dispensings in two ways: (1) at any point after diagnosis to capture all use, and (2) in the two-year period following diagnosis to approximate adjuvant use.

RESULTS

Over time, adjuvant aromatase inhibitor use increased whereas tamoxifen use decreased. Aromatase inhibitor dispensings within 2 years of diagnosis increased from 4.1% among women diagnosed in 2000 to 13% in 2001, 24% in 2002, and 40% in 2003. Tamoxifen use declined starting in 2001 at every system.

CONCLUSION

Aromatase inhibitor use rose dramatically after 2001 while tamoxifen use decreased. It appears results from early clinical trials changed practice in these integrated healthcare systems before formal changes in national guidelines.